FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE

Big Data and Machine Learning in NETL’s Fossil Energy PortfolioSolutions for Today | Options for Tomorrow

Randall Gentry, Ph.D.Chief Research Officer

July 12, 2018

FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE 2

Core Competencies & FE Technology Thrusts

Materials Engineering & Manufacturing

Geological & Environmental

Systems

Energy Conversion Engineering

Systems Engineering & Analysis

ComputationalScience & Engineering

Program Execution & Integration

MethaneHydrates

EnhancedResource Production

EnvironmentallyPrudent Development

Sensors & Controls

OIL &

GAS

COAL

CarbonStorage

CarbonCapture

AdvancedMaterials

Advanced EnergySystems

AdvancedComputing

Water Management

Rare Earth Elements

Offshore UnconventionalNatural GasInfrastructure

Microgrid Energy Security & RestorationEnergy StorageVehicles Solid State Lighting Geothermal

Energy Efficiency & Renewable Energy (EERE) Electricity Delivery & Energy Reliability (OE)Support to Other

DOE Offices

FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE 3

FE Technology Thrust Integration

OIL & GAS PROGRAM

Midstream

InfrastructureMethane

Quantification

Methane

Hydrates

Rare Earth

Elements

Carbon

StorageCarbon

Capture

Crosscutting

Research &

Analysis

Advanced

Energy

Systems

STEP

(Supercritical

CO2)

Offshore

Revitalize and Extend Coal

Enable CO2 as a Commodity for Energy Security

Modernize existing coal plants

Develop & Deploy Next Gen Coal-Based Energy Systems

Get the Most out of Coal Resources

COAL INITIATIVES

OIL & GAS INITIATIVES

Improved Recovery Efficiency of Resources

Advanced Resource Characterization

Natural Gas Utilization Development

Natural Gas Infrastructure Improvement

Reduced Operational Impacts

Grow Oil & Gas

Impact

Manufacturing Revitalization

Global Competitiveness

Energy Dominance

Economic Growth

Unconventional

Oil & Gas

COAL PROGRAM

FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE 4

Fossil-focus in Advanced Manufacturing

Big Data and Machine Learning to improve the

performance and economics of energy and

materials systems

IMPACT

The advanced coal energy systems of the future:Create new long-term pathways for advanced coal

energy (ACE) systems, supported by the most

advanced and innovative technologies;

A competitive, resilient and flexible fleet:Identify ways to strengthen and utilize existing plants

that would provide affordable near-term energy

security benefits and also support future power and

infrastructure needs amidst a changing energy

landscape; and

New Markets:Develop new products and uses of coal and coal

by-products to create new businesses and industries.

NETL collaborates in three of the Manufacturing USA Institutes: America Makes, RAPID, and ARM.

Nano-manufacturing

Artificial Intelligence

Machine Learning

Additive Manufacturing (America Makes)

Cyber-Physical

Robotics(ARM)

Ensuring the economic vitality of coal at the

intersection of energy and advanced manufacturing

Modular Process Intensification

(RAPID)

Big Data

5

Active Portfolio Leveraging Big Data and Machine Learning

Predictive Maintenance

Digital Twinning

Sensors and Controls

IDAES

CCSI2

HPC4CM

Subsurface

FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE 6

Predictive Maintenance

• Microbeam Technologies, Inc.Integrated Predictive Maintenance to integrate the operations of the

tool into plant control systems and plant operating parameters. These

improvements will potentially allow automation of coal selection and

blending and will enhance the efficiency and long-term reliability of

coal plants.

• SparkCognitionThe approach utilizes existing sensor and operational data being

collected at coal-fired power plants and apply its machine-learning

algorithms to detect and diagnose premature equipment failure.

Benefits from successful completion of this project include optimizing

the sensor inputs needed for fault detection, understanding the

impacts of control decisions due to flexible operations, and extending

the life of critical equipment.

Predictive Maintenance at NETL

Utilizing data from distributed sensors and applying

machine learning to diagnose faults before they

occur will lead to:

• Converting from a culture of preventative

maintenance to one of condition-based

maintenance.

• Identifying operational discontinuities and

informing decisions on operational efficiency

• Enabling plants to operate in an environment

where they are required to cycle more

frequently than originally envisioned

Key Outcomes

FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE 7

Digital Twinning

• JOULE coal plant digital twinsUtilize coal plant data to produce Digital Twin models including all necessary aspects of the physical asset or larger system such as thermal, mechanical, electrical,

chemical, fluid dynamic, material, lifing, economic and statistical. These models also accurately represent the plant or fleet under a large number of variations related to operation — fuel mix, ambient temperature, air quality, moisture, load, weather forecast models, and marketpricing.

Digital Twinning at NETL

Compliments machine learning by developing a

twin computational model to understand and

predict the impact of change to the real world

application:

• Cost savings

• Risk mitigation

• Safety and security improvements

Key Outcomes

Big Data

Prescriptive Analytics

FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE 8

Sensors and Controls

• Real-time measurement of temperature profiles in

boiler different combustion zones

• Detection of target gases at high temperatures and

electrochemical sensors

• Wireless Condition-

Based Monitoring

• Distributed Fiber

Optic Sensing Systems

Sensors and Controls at NETL

Advancing sensors and controls with material

improvements, algorithm development,

data-driven hybrid models integrated into

the central controls, and application of

advanced control systems (including

distributed intelligence):

• Increase coal plant efficiency

• Reduce forced outages

• Safety and security improvements

Key Outcomes

FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE 9FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE

Institute for the Design of Advanced Energy Systems (IDAES)

• Institute for the Design of Advanced Energy

Systems (IDAES)IDAES team has implemented a modular framework and model library that supports large-scale optimization of advanced energy systems; applied machine learning-based parameter estimation tools; developed a roadmap to support the existing fleet of coal-fired power plants; and established an industry stakeholder advisory board.

IDAES at NETL

The Institute is a resource for the development and

analysis of innovative advanced energy systems

via process systems engineering tools and

approaches. IDAES benefits are:

• Process Synthesis, Integration, and Intensification

• Process Control and Dynamics

• Apply to development of novel energy systems

• Transformational Carbon Capture

• National Lab and University Capability

• Open Source

Key Outcomes

FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE 10FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE

Rational Design for Solvents and for carbon capture experiments (CCSI2)

• CCSI2The CCSI Toolset is designed provide end users in industry with a comprehensive, integrated suite of scientifically validated models, with uncertainty quantification, optimization, risk analysis and decision making capabilities.

The CCSI Toolset incorporates commercial and open-source software currently in use by industry and is also developing new software tools as necessary to fill technology gaps identified during execution of the project. The current focus is on using machine learning and data from past pilot projects to optimally design experiments for carbon capture

CCSI2 at NETL

The Carbon Capture Simulation Initiative (CCSI) is a

partnership among national laboratories, industry,

and academic institutions that is developing and

deploying state-of-the-art computational modeling

and simulation tools to accelerate the

commercialization of carbon capture

technologies. CCSI2 Toolkit benefits are:

• Prediction of coal quality in operations

• Carbon capture modeling

• Advanced process simulation

Key Outcomes

FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE 11FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE

High Performance Materials Development (HPC4M)

• HPC4CMNETL is focused on improving the existing coal fleet by characterizing, producing, and certifying high performance materials for use in extreme environments. To do this, NETL focuses on four areas of research in materials: computational materials design, advanced structural materials, functional materials for process performance, and advanced manufacturing techniques.

HPC4CM at NETL

Through the high performance computing for

manufacturing (HPC4M) program, key challenges

in developing, modifying and qualifying new

materials are being advanced using machine

learning and big data:

• Accelerates new material identification

• Advanced material properties

Key Outcomes

Computational

Materials Design

Advanced

Structural

Materials

Functional

Materials for

Process

Performance

Advanced

Manufacturing

FOR INTERNAL USE ONLY – NOT FOR PUBLIC RELEASE 12

Subsurface

• SubsurfaceSuccessfully engineering the subsurface requires advanced quantitative assessment and characterization of the geologic strata, tools, and materials used to access and image the deep subsurface, as well as computational tools

required to analyze significant volumes of data and model complex coupled reactions.

Subsurface at NETL

Research, scientific, and engineering data

resource are increasingly available online. For the

subsurface, these resources span a tremendous

amount of data, models and analysis. Energy Data

eXchange (EDX), helps improve access to these

resources by:

• Accelerates coordination and access

• Advanced search capabilities

• Potential for big data and machine learning

Key Outcomes

13

What’s Next?

• Improvements in

program management and stewardship through machine intelligence

• Creating a more robust portfolio and ensuring program success

• Using smart

manufacturing to bring down costs

• Leverage integrated sensors and control systems to increase plant efficiency, understand component

health, and improve environmental performance

• Robotic technology to enable automatic plant inspection and repairs

• Robotic technologies provide non-

destructive testing inspections ready for commercial applications

• Develop that enhance the cybersecurity of advanced sensor and control networks

• Enhance plant flexibility with secure sensor data

transmission

• Cutting edge research projects aim to reduce operating and maintenance costs to make powerplants economically competitive

• Example project includes development of large-diameter, multi-nozzle turbine combustors

BLOCKCHAIN &

CYBERSECURITY

MACHINE

LEARNING FOR

PROGRAM

MANAGEMENT

SENSORS AND

CONTROLS

ROBOTICS

ENABLED

TECHNOLOGIES

POWERPLANT

AUTOMATION

Thank

You.

MORGANTOWN, WV3610 Collins Ferry RoadP.O. Box 880Morgantown, WV 26507-0880304-285-4764

CONTACTU.S. Department of EnergyNational Energy Technology Laboratory

PITTSBURGH, PA626 Cochrans Mill RoadP.O. Box 10940Pittsburgh, PA 15236-0940412-386-4984

ALBANY, OR1450 Queen Avenue SWAlbany, OR 97321-2198541-967-5892